
#######################################################################


import pandas as pd
import matplotlib.pyplot as plt
import plotly.express as px
import math
import numpy as np


df = pd.read_csv("co2_emission_by_countries.csv", encoding="latin")
df1 = df.copy()
df = pd.read_csv("continent_data.csv")
df2 = df.copy()


df1.describe()


df1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 59620 entries, 0 to 59619
Data columns (total 9 columns):
 #   Column               Non-Null Count  Dtype  
---  ------               --------------  -----  
 0   Country              59620 non-null  object 
 1   Code                 57452 non-null  object 
 2   Calling Code         56097 non-null  object 
 3   Year                 59620 non-null  int64  
 4   CO2 emission (Tons)  59620 non-null  float64
 5   Population(2022)     53116 non-null  float64
 6   Area                 55284 non-null  float64
 7   % of World           55284 non-null  object 
 8   Density(km2)         53116 non-null  object 
dtypes: float64(3), int64(1), object(5)
memory usage: 4.1+ MB


df1.shape

(59620, 9)


df1['Country'].value_counts().unique()

array([271], dtype=int64)


df1 = df1.drop(columns=['Calling Code','Code'])

df1


df1 = df1.rename(columns={'CO2 emission (Tons)': 'coTons', 'Population(2022)' : 'population'})

df1


df1 = df1.sort_values('population', ascending = False)


dfSelect2020 = df1[df1['Year'] == 2020]


dfSelect2020


dfNull2020 = dfSelect2020[dfSelect2020.isnull().any(axis=1)]


dfNull2020['Country']

23576                       Hong Kong
41191                       Palestine
39836                 North Macedonia
27912                          Kosovo
43901                         Reunion
9755                       Cape Verde
7858                           Brunei
1625                         Anguilla
1896                       Antarctica
2980                            Aruba
6232                          Bermuda
11110                Christmas Island
11923                           Congo
12194                    Cook Islands
12736                   Cote d'Ivoire
13549                         Curacao
14091                         Czechia
14362    Democratic Republic of Congo
17885                  Faeroe Islands
19240                French Polynesia
22492                   Guinea-Bissau
31164                           Macao
35771                      Montserrat
37939                   New Caledonia
39294                            Niue
44985                    Saint Helena
45798       Saint Pierre and Miquelon
48508       Sint Maarten (Dutch part)
53657                           Timor
55554        Turks and Caicos Islands
58806               Wallis and Futuna
Name: Country, dtype: object


len(dfNull2020.index)

31


the31Union = dfNull2020['coTons'].sum()
theMostPoluating = dfSelect2020.nlargest(1, 'coTons')


x_value = [theMostPoluating['Country'].values[0], 'U34']
y_value = [theMostPoluating['coTons'].values[0], the31Union]

plt.bar( x_value, y_value)

plt.show()


dfSorted2020 = dfSelect2020.sort_values('coTons', ascending=True)
dfSorted2020


i = 0
countriesSumsCo2 = 0
dfCountriesSums = pd.DataFrame(columns=df1.columns)


while theMostPoluating['coTons'].values[0] > countriesSumsCo2 :
    
    countriesSumsCo2 += dfSorted2020['coTons'].values[i] 

    row = pd.DataFrame([dfSorted2020.iloc[i]])
    dfCountriesSums = pd.concat([dfCountriesSums, row])

    
    i += 1
print(str(i) + " " + str(countriesSumsCo2))

207 418617496416.0


dfCountriesSums


df2020Big = dfSelect2020.nlargest(10, 'coTons')


df2020Big


allYearCO2Big = df1[df1['Country'].isin(df2020Big['Country'])]


allYearCO2Big


pivot_table = allYearCO2Big.pivot_table(values='coTons', index='Year', columns='Country')


pivot_table.plot()
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


dfCentury19 = allYearCO2Big[(allYearCO2Big['Year'] >= 1800) & (allYearCO2Big['Year'] <= 1900)]


pivot_table = dfCentury19.pivot_table(values='coTons', index='Year', columns='Country')


pivot_table.plot()
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


dfCentury20 = allYearCO2Big[(allYearCO2Big['Year'] >= 1900) & (allYearCO2Big['Year'] <= 2000)]


pivot_table = dfCentury20.pivot_table(values='coTons', index='Year', columns='Country')


pivot_table.plot()
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


fig = px.line(pivot_table, x=pivot_table.index, y=pivot_table.columns[0:])
fig.show()


dfCentury21 = allYearCO2Big[(allYearCO2Big['Year'] >= 2000) & (allYearCO2Big['Year'] <= 2020)]


pivot_table = dfCentury21.pivot_table(values='coTons', index='Year', columns='Country')


pivot_table


pivot_table.plot()
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


pivot_table.plot(figsize=(15, 5), rot=30)
plt.xticks(ticks=pivot_table.index, labels=pivot_table.index.astype(int))
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


df2020BigByArea = df2020Big
df2020BigByArea['coByArea'] = (df2020BigByArea['coTons'] / df2020BigByArea['Area'])


df2020BigByArea


df2020BigByArea.plot.bar(x='Country', y='coByArea', rot=30, figsize=(10,5))
plt.show()


allYearCoAreaBig = allYearCO2Big.copy()
allYearCoAreaBig['coByArea'] = (allYearCoAreaBig['coTons'] / allYearCoAreaBig['Area'])


allYearCoAreaBig


pivot_table = allYearCoAreaBig.pivot_table(values='coByArea', index='Year', columns='Country')


pivot_table.plot()
plt.ylabel('CO2 TONS')
plt.title('CO2 Emissions over time')
plt.show()


fig = px.line(pivot_table, x=pivot_table.index, y=pivot_table.columns[0:])


fig.show()


df1['co2ForYear'] = math.nan
df1 = df1.sort_values(['Country','Year'], ascending=[True,True])
df1['co2ForYear'] = df1.groupby('Country')['coTons'].diff()
df1['co2ForYear'] = df1['co2ForYear'].fillna(df1['coTons'])

df1


mostCoByYear = df1[df1['Country'].isin(df2020Big['Country'].values)]


mostCoByYear


pivot_table = mostCoByYear.pivot_table(index = 'Year', values='co2ForYear',columns='Country' )


fig = px.line(pivot_table, x=pivot_table.index, y=pivot_table.columns[0:])
fig.show()


df2test = df2[df2['Entity'].str.contains('united', case=False)]
df2test


df1['Continent'] = math.nan


df1.head(10)


df1[df1['Country'].isin(df2['Entity'])]


df1[~df1['Country'].isin(df2['Entity'])]['Country'].unique()

array(['Faeroe Islands', 'Micronesia', 'Timor'], dtype=object)


concat = pd.DataFrame({'Entity':['Faeroe Islands','Micronesia','Timor'], 'Code': math.nan, 'Year': math.nan, 'Continent':['Europe', 'Oceania', 'Asia']} )


df2 = pd.concat([df2, concat])

df2


df2.drop(['Code', 'Year'], axis=1, inplace=True)


df2.head()


df2.reset_index(drop=True, inplace=True)


df1['Country'].dtype

dtype('O')


df2_dic = df2.set_index('Entity')['Continent'].to_dict()
df1['Continent'] = df1['Country'].map(df2_dic)


df1.groupby('Country').first().head(10)


def getPlot (df,category,type, plotType, mark):
        
    if plotType == 'bar':
        
            
        #sum value for getting value for continent 
        if category == 'Continent':
            df = df.groupby(['Year',category])[type].sum().reset_index()
        
        fig = px.bar(df, x='Year', y=type, color = category, barmode='group',  title=(type + ' By ' + category))
        fig.show()

    
    if plotType == 'line':
        fig = px.line(df, x=df.index, y=df.columns[0:], markers = mark,  title=(type + ' By ' + category))
        
        fig.show()


def getPivot(df, category, type):
    pivot_table = None

    #pivot table with sum of all '$type' have the same '$category' 
    if category == 'Continent':
        pivot_table = df.pivot_table(index='Year', columns=category, values=type, aggfunc='sum')

        
    else:
        pivot_table = df.pivot_table(index='Year', columns=category, values=type)
    
    return pivot_table


def getFilter(year,  category, range, list, df):

        # set on first letter for prevent error
        list = [item.capitalize() for item in list]

        #set a dataframe for a range of value
        if range == 'Yes' :
            df_filtered = df[df[category].isin(list) & df['Year'].between(year[0], year[1])]  

        #set a dataframe for a range of value take one or many value
        else :
            df_filtered = df[df[category].isin(list) & df['Year'].isin(year)]
            
        return df_filtered


"""

Year -> take a list of date to select one or many
Category -> If you need country data or ContinentData
Range -> take string if Yes : return year in a range, else : selected year only
df -> taking the dataframe 
list -> list of country
type -> take a list of continent or country
plotType ->'bar' or 'line'
marker if line is True
"""

"\n\nYear -> take a list of date to select one or many\nCategory -> If you need country data or ContinentData\nRange -> take string if Yes : return year in a range, else : selected year only\ndf -> taking the dataframe \nlist -> list of country\ntype -> take a list of continent or country\nplotType ->'bar' or 'line'\nmarker if line is True\n"


def renderPlot(year, category, range, list, df,type, plotType, mark):

    category = category.capitalize()
    
    filterData = getFilter(year, category, range, list, df)
      
    if plotType == 'line':
        BernardPivot = getPivot(filterData, category, type )
        return getPlot(BernardPivot,category,type, plotType, mark)

    else:
        return getPlot(filterData,category,type, plotType, mark)


renderPlot([1995,1998, 2000],'Country', 1, ['France', 'Belgium','Yemen'], df1,'coTons', 'line', True)


renderPlot([1995, 2002, 2000],'Continent', 'Yes', ['europe', 'asia'], df1,'coTons', 'bar', True)


renderPlot([1995, 2002],'Continent', 'Yes', ['europe', 'asia'], df1,'coTons', 'line', True)

	Year	CO2 emission (Tons)	Population(2022)	Area
count	59620.000000	5.962000e+04	5.311600e+04	5.528400e+04
mean	1885.000000	1.034774e+09	3.992260e+07	6.522073e+05
std	78.231085	1.041652e+10	1.482365e+08	1.865483e+06
min	1750.000000	0.000000e+00	1.131200e+04	2.100000e+01
25%	1817.000000	0.000000e+00	1.770414e+06	1.770450e+04
50%	1885.000000	0.000000e+00	8.673095e+06	1.103815e+05
75%	1953.000000	8.715092e+06	2.862920e+07	4.925730e+05
max	2020.000000	4.170000e+11	1.425887e+09	1.709824e+07

	Country	Year	CO2 emission (Tons)	Population(2022)	Area	% of World	Density(km2)
0	Afghanistan	1750	0.0	41128771.0	652230.0	0.40%	63/km²
1	Afghanistan	1751	0.0	41128771.0	652230.0	0.40%	63/km²
2	Afghanistan	1752	0.0	41128771.0	652230.0	0.40%	63/km²
3	Afghanistan	1753	0.0	41128771.0	652230.0	0.40%	63/km²
4	Afghanistan	1754	0.0	41128771.0	652230.0	0.40%	63/km²
...	...	...	...	...	...	...	...
59615	Zimbabwe	2016	736467042.0	16320537.0	390757.0	0.30%	42/km²
59616	Zimbabwe	2017	746048675.0	16320537.0	390757.0	0.30%	42/km²
59617	Zimbabwe	2018	757903042.0	16320537.0	390757.0	0.30%	42/km²
59618	Zimbabwe	2019	768852126.0	16320537.0	390757.0	0.30%	42/km²
59619	Zimbabwe	2020	779383468.0	16320537.0	390757.0	0.30%	42/km²

	Country	Year	coTons	population	Area	% of World	Density(km2)
0	Afghanistan	1750	0.0	41128771.0	652230.0	0.40%	63/km²
1	Afghanistan	1751	0.0	41128771.0	652230.0	0.40%	63/km²
2	Afghanistan	1752	0.0	41128771.0	652230.0	0.40%	63/km²
3	Afghanistan	1753	0.0	41128771.0	652230.0	0.40%	63/km²
4	Afghanistan	1754	0.0	41128771.0	652230.0	0.40%	63/km²
...	...	...	...	...	...	...	...
59615	Zimbabwe	2016	736467042.0	16320537.0	390757.0	0.30%	42/km²
59616	Zimbabwe	2017	746048675.0	16320537.0	390757.0	0.30%	42/km²
59617	Zimbabwe	2018	757903042.0	16320537.0	390757.0	0.30%	42/km²
59618	Zimbabwe	2019	768852126.0	16320537.0	390757.0	0.30%	42/km²
59619	Zimbabwe	2020	779383468.0	16320537.0	390757.0	0.30%	42/km²

	Country	Year	coTons	population	Area	% of World	Density(km2)
10839	China	2020	2.360000e+11	1.425887e+09	9706961.0	6.30%	147/km²
24389	India	2020	5.441649e+10	1.417173e+09	3287590.0	2.00%	431/km²
57180	United States	2020	4.170000e+11	3.382899e+08	9372610.0	6.10%	36/km²
24660	Indonesia	2020	1.439996e+10	2.755013e+08	1904569.0	1.20%	145/km²
40649	Pakistan	2020	5.162289e+09	2.358249e+08	881912.0	0.50%	267/km²
...	...	...	...	...	...	...	...
45798	Saint Pierre and Miquelon	2020	3.676592e+06	NaN	242.0	0.00%	NaN
48508	Sint Maarten (Dutch part)	2020	6.770353e+07	NaN	NaN	NaN	NaN
53657	Timor	2020	6.543008e+06	NaN	NaN	NaN	NaN
55554	Turks and Caicos Islands	2020	3.930547e+06	NaN	948.0	0.00%	NaN
58806	Wallis and Futuna	2020	5.072470e+05	NaN	NaN	NaN	NaN

	Country	Year	coTons	population	Area	% of World	Density(km2)
1896	Antarctica	2020	1.538880e+05	NaN	NaN	NaN	NaN
43359	Puerto Rico	2020	2.088480e+05	3.252407e+06	8870.0	0.00%	367/km²
55825	Tuvalu	2020	2.789250e+05	1.131200e+04	26.0	0.00%	435/km²
39294	Niue	2020	2.901480e+05	NaN	260.0	0.00%	NaN
44985	Saint Helena	2020	3.792950e+05	NaN	394.0	0.00%	NaN
...	...	...	...	...	...	...	...
56909	United Kingdom	2020	7.816115e+10	6.750894e+07	242900.0	0.20%	278/km²
20324	Germany	2020	9.263562e+10	8.336984e+07	357114.0	0.20%	233/km²
44443	Russia	2020	1.150000e+11	1.447133e+08	17098242.0	11.00%	8/km²
10839	China	2020	2.360000e+11	1.425887e+09	9706961.0	6.30%	147/km²
57180	United States	2020	4.170000e+11	3.382899e+08	9372610.0	6.10%	36/km²

Study of Co2 from 1700 to 2020 by country¶

Imports librairies¶

import data from csv¶

reading data as dataframe¶

Remove usless columns as "Calling code" and "Code"¶

Rename columns of select easier¶

Sorting the different columns to find correlations or anomalies¶

as we can see, we have a lot of NaN value let's see in 2020 how many Country contain NaN information in this data set¶

create filter for geting all data where year is 2020¶

create filter for return all rows where one value have NaN on all columns¶

how many are them ?¶

pure curiosity, let's compare the most polluting to this group of countries¶

I'm curious to find out the number of countries required to match the CO2 emissions of the United States¶

After observing this, i understand that CO2 emissions are cumulative from 1700 rather than being based annualy starting from 1700.¶

and when we see the united state at 417 billion tons co2Tons that confirm the data 'coTons' is cumulative and not based annualy¶

let's get 10 country who generated the most of c02 in 2020¶

Getting data from all year for the 10 selected¶

creating pivot_table for setting year as index and country as column for Emission though year on graph¶

Let's see now how it was on the 19th century¶

Let's see now how it was on the 20th century¶

Let's see now how it was on the 21th century¶

as we can see i have a problem on my X axis¶

now we fixed the issue, let's go to the next step¶

well let's now see from this ten coutry who's generated most co2 by Area we gonna take 2020 data for the first step¶

let's get coByArea from 1700¶

create the pivot table for getting the graph¶

i want to have a better vision with a dynamic graph, let's use plotly¶

manipulate data for transform Cumulated CO2 from 1750 to Year by Year CO2¶

let's take the same country as before for see how it is for each year¶

As we can see, the measurements provided by the dataset are not precise enough. We will conclude with a comparison by continent and a prediction model for the coming years.¶

Let's manipulate data continent¶

as we can see we have 3 country missing¶

Mapping for each country he's Continent¶

Country	Canada	China	France	Germany	India	Japan	Russia	Ukraine	United Kingdom	United States
Year
2000	2.218015e+10	7.488403e+10	3.142863e+10	7.621667e+10	1.893972e+10	4.100939e+10	8.332305e+10	2.494736e+10	6.847098e+10	3.040000e+11
2001	2.273920e+10	7.839897e+10	3.183916e+10	7.713332e+10	1.993228e+10	4.225916e+10	8.483076e+10	2.525092e+10	6.904729e+10	3.100000e+11
2002	2.330387e+10	8.227156e+10	3.224517e+10	7.803329e+10	2.095531e+10	4.353828e+10	8.632645e+10	2.554660e+10	6.960736e+10	3.160000e+11
2003	2.388621e+10	8.681658e+10	3.265712e+10	7.893444e+10	2.201493e+10	4.482575e+10	8.785206e+10	2.585361e+10	7.017883e+10	3.220000e+11
2004	2.446697e+10	9.204033e+10	3.307027e+10	7.982153e+10	2.314040e+10	4.610861e+10	8.938275e+10	2.616398e+10	7.075173e+10	3.280000e+11
2005	2.504282e+10	9.791689e+10	3.348578e+10	8.068823e+10	2.432635e+10	4.739898e+10	9.093035e+10	2.647710e+10	7.132169e+10	3.340000e+11
2006	2.561423e+10	1.040000e+11	3.389154e+10	8.156655e+10	2.558610e+10	4.866634e+10	9.253690e+10	2.680977e+10	7.188926e+10	3.400000e+11
2007	2.620889e+10	1.110000e+11	3.428697e+10	8.241817e+10	2.694425e+10	4.996948e+10	9.414145e+10	2.714613e+10	7.244850e+10	3.470000e+11
2008	2.678826e+10	1.190000e+11	3.467496e+10	8.327310e+10	2.840706e+10	5.120179e+10	9.577837e+10	2.747166e+10	7.299300e+10	3.530000e+11
2009	2.733487e+10	1.270000e+11	3.504359e+10	8.406339e+10	3.001988e+10	5.236500e+10	9.730733e+10	2.774896e+10	7.348689e+10	3.580000e+11
2010	2.789368e+10	1.350000e+11	3.542017e+10	8.489634e+10	3.169777e+10	5.357982e+10	9.892044e+10	2.804303e+10	7.399853e+10	3.640000e+11
2011	2.846355e+10	1.450000e+11	3.577391e+10	8.570556e+10	3.347790e+10	5.484469e+10	1.010000e+11	2.835099e+10	7.446798e+10	3.690000e+11
2012	2.903281e+10	1.550000e+11	3.612984e+10	8.651954e+10	3.544148e+10	5.615070e+10	1.020000e+11	2.865496e+10	7.495550e+10	3.750000e+11
2013	2.960651e+10	1.650000e+11	3.648737e+10	8.735100e+10	3.747842e+10	5.746604e+10	1.040000e+11	2.895221e+10	7.543288e+10	3.800000e+11
2014	3.017707e+10	1.750000e+11	3.681326e+10	8.814359e+10	3.966428e+10	5.872977e+10	1.060000e+11	2.920972e+10	7.587142e+10	3.860000e+11
2015	3.075013e+10	1.840000e+11	3.714317e+10	8.893920e+10	4.193284e+10	5.995316e+10	1.070000e+11	2.943354e+10	7.629376e+10	3.910000e+11
2016	3.130970e+10	1.940000e+11	3.747675e+10	8.973988e+10	4.431507e+10	6.115688e+10	1.090000e+11	2.966755e+10	7.669310e+10	3.960000e+11
2017	3.187906e+10	2.040000e+11	3.781374e+10	9.052577e+10	4.674892e+10	6.234500e+10	1.100000e+11	2.989064e+10	7.708067e+10	4.010000e+11
2018	3.245853e+10	2.140000e+11	3.813612e+10	9.127988e+10	4.934873e+10	6.348846e+10	1.120000e+11	3.012267e+10	7.746256e+10	4.070000e+11
2019	3.304092e+10	2.250000e+11	3.845205e+10	9.199130e+10	5.197470e+10	6.459439e+10	1.140000e+11	3.034525e+10	7.783157e+10	4.120000e+11
2020	3.357674e+10	2.360000e+11	3.872868e+10	9.263562e+10	5.441649e+10	6.562517e+10	1.150000e+11	3.055916e+10	7.816115e+10	4.170000e+11

	Country	Year	coTons	population	Area	% of World	Density(km2)	co2ForYear
9214	Canada	1750	0.000000e+00	38454327.0	9984670.0	6.10%	4/km²	0.000000e+00
9215	Canada	1751	0.000000e+00	38454327.0	9984670.0	6.10%	4/km²	0.000000e+00
9216	Canada	1752	0.000000e+00	38454327.0	9984670.0	6.10%	4/km²	0.000000e+00
9217	Canada	1753	0.000000e+00	38454327.0	9984670.0	6.10%	4/km²	0.000000e+00
9218	Canada	1754	0.000000e+00	38454327.0	9984670.0	6.10%	4/km²	0.000000e+00
...	...	...	...	...	...	...	...	...
57176	United States	2016	3.960000e+11	338289857.0	9372610.0	6.10%	36/km²	5.000000e+09
57177	United States	2017	4.010000e+11	338289857.0	9372610.0	6.10%	36/km²	5.000000e+09
57178	United States	2018	4.070000e+11	338289857.0	9372610.0	6.10%	36/km²	6.000000e+09
57179	United States	2019	4.120000e+11	338289857.0	9372610.0	6.10%	36/km²	5.000000e+09
57180	United States	2020	4.170000e+11	338289857.0	9372610.0	6.10%	36/km²	5.000000e+09

	Entity	Code	Year	Continent
263	United Arab Emirates	ARE	2015	Asia
264	United Kingdom	GBR	2015	Europe
265	United States	USA	2015	North America
266	United States Minor Outlying Islands	UMI	2015	Oceania
267	United States Virgin Islands	VIR	2015	North America

	Entity	Code	Year	Continent
0	Abkhazia	OWID_ABK	2015.0	Asia
1	Afghanistan	AFG	2015.0	Asia
2	Akrotiri and Dhekelia	OWID_AKD	2015.0	Asia
3	Aland Islands	ALA	2015.0	Europe
4	Albania	ALB	2015.0	Europe
...	...	...	...	...
283	Zanzibar	OWID_ZAN	2015.0	Africa
284	Zimbabwe	ZWE	2015.0	Africa
0	Faeroe Islands	NaN	NaN	Europe
1	Micronesia	NaN	NaN	Oceania
2	Timor	NaN	NaN	Asia